The purpose of this study was to develop and validate an individualized nomogram to predict venous thromboembolism (VTE) risk in hospitalized postoperative breast cancer patients.A single-central retrospective and non-interventional trial.For model development, we used data from 4,755 breast cancer patients between 1 November 2016-30 June 2018 (3,310 patients in the development group and 1,445 in the validation group). Overall, 216 patients developed VTE (150 in development group and 66 in validation group). The model was validated by receiver operating characteristic curves and the calibration plot. The clinical utility of the model was determined through decision curve analysis.The individualized nomogram consisted of six clinical factors: age, body mass index, number of cardiovascular comorbidities, neoadjuvant chemotherapy, surgical treatment, hospital length of stay and two pre-operative biomarkers of Homocysteine and D-dimer. The model at the 3.9% optimal cut-off had the area under the curve of 0.854 (95% CI, 0.824-0.884) and 0.805 (95% CI, 0.740-0.870) in the development and validation groups. A p = 0.570 of the calibration test showed that the model was well-calibrated. The net benefit of the model was better between threshold probabilities of 5%-30% in decision curve analysis.The nomogram of VTE risk assessment, is applicable to hospitalized postoperative breast cancer patients. However, multi-central prospective studies are needed to improve and validate the model. Effectiveness and safety of thromboprophylaxis in high-risk patients are needed to demonstrate in interventional trials.This nomogram can be used in clinical to inform practice of physicians and nurses to predict the VTE probability and maybe direct personalized decision making for thromboprophylaxis in hospitalized postoperative breast cancer patients.目标: 本研究的目的是开发和验证个体化列线图,以预测乳腺癌术后住院患者静脉血栓栓塞(VTE)的风险。 设计: 单中心回顾性非干预试验。 方法: 对于模型开发,我们使用了2016年11月1日至2018年6月30日期间4755名乳腺癌患者的数据(发生组3310名,验证组1445名)。总体来说,216例患者发生了静脉血栓栓塞(发生组150例,验证组66例)。通过接收者工作特性曲线和标定曲线对模型进行验证。通过决策曲线分析确定模型的临床实用性。 结果: 个体化列线图由6个临床因素组成:年龄、体重指数、心血管共病数量、新辅助化疗、手术治疗、住院时间以及术前同型半胱氨酸和D-二聚体两个生物标志物。在开发组和验证组中,3.9%最优截止点的模型下面积为0.854(95%可信区间,0.824-0.884)和0.805(95%可信区间,0.740-0.870)。校准试验的p = 0.570 表明模型校准良好。在决策曲线分析中,阈值概率在5%-30%之间,模型的净效益较好。 结论: 静脉血栓栓塞风险评估列线图适用于乳腺癌术后住院患者。然而,需要多中心前瞻性研究来改进和验证该模型。高危患者血栓预防的有效性和安全性需要在干预试验中证明。 影响: 该列线图可用于临床指导医生和护士预测静脉血栓栓塞发生率,并可指导乳腺癌术后住院患者血栓预防的个性化决策。.